The Role of Thoracic Radiotherapy as an Adjunct to Standard Chemotherapy in Limited-Stage Small Cell Lung Cancer

Transcription

1 Evidence-based Series # EDUCATION AND INFORMATION 2013 The Role of Thoracic Radiotherapy as an Adjunct to Standard Chemotherapy in Limited-Stage Small Cell Lung Cancer Members of the Lung Cancer Disease Site Group A Quality Initiative of the Program in Evidence-based Care (PEBC), Cancer Care Ontario (CCO) An assessment conducted in November 2013 put Evidence-based Series (EBS) in the Education and Information section. This means that the recommendations will no longer be maintained by may still be useful for academic or other information purposes. The PEBC has a formal and standardized process to ensure the currency of each document (PEBC Assessment & Review Protocol). The reviewed EBS report, consists of Section 1: Clinical Practice Guideline Section 2: Systematic Review Section 3: Document Review Summary and Tool and is available on the CCO Web site ( PEBC Lung Cancer DSG page at: Release Date: May For information about the PEBC and the most current version of all reports, please visit the CCO website at or contact the PEBC office at: Phone: ext Fax: ccopgi@mcmaster.ca Guideline Citation (Vancouver Style): Members of the Lung Cancer Disease Site Group. The role of thoracic radiotherapy as an adjunct to standard chemotherapy in limited-stage small cell lung cancer. Toronto (ON): Cancer Care Ontario; 2003 Jan [EDUCATION AND INFORMATION 2013]. Program in Evidence-based Care Practice Guideline Report No.: EDUCATION AND INFORMATION 2013

2 Guideline Report History GUIDELINE VERSION Search Dates SYSTEMATIC REVIEW Data PUBLICATIONS NOTES AND KEY CHANGES Original version October Full Report Peer-reviewed publication 1 Web publication NA d version January New data added to original full report d Web publication 1999 recommendations revised and modified with new evidence Reviewed Version May New data found in section 3: Document Review Summary and Tool d Web publication 2003 recommendations require an UPDATE 1 Okawara G, Gagliardi A, Evans WK, and the Cancer Care Ontario Practice Guidelines Initiative Lung Cancer Disease Site Group. The role of thoracic radiotherapy as an adjunct to standard chemotherapy in limited-stage small-cell lung cancer. Curr Oncol 2000;7(3):

3 Evidence-based Series # : Section 1 The Role of Thoracic Radiotherapy as an Adjunct to Standard Chemotherapy in Limited-Stage Small Cell Lung Cancer: A Clinical Practice Guideline G. Okawara, A. Gagliardi, W. K. Evans, and members of the Lung Cancer Disease Site Group A Quality Initiative of the Program in Evidence-Based Care (PEBC), Cancer Care Ontario (CCO) ORIGINAL GUIDELINE: October 8, 1999 RECOMMENDATIONS MODIFIED: January 2003 MOST RECENT LITERATURE SEARCH: December 2002 NEW EVIDENCE ADDED TO GUIDELINE REPORT: January 2003 Based on new evidence that emerged after completion of the original guideline, the Lung Cancer Disease Site Group modified the guideline recommendations in January The revised recommendations are labelled. Report Date: January 2003 Guideline Question Is there a role for thoracic radiotherapy as an adjunct to standard chemotherapy in limited-stage small cell lung cancer? Target Population These recommendations apply to adult patients with limited-stage small cell lung cancer. Recommendations In patients with limited-stage small cell lung cancer, the addition of thoracic radiotherapy to standard combination chemotherapy improves both local control and overall survival and should be incorporated into a comprehensive treatment plan of combined modality therapy for limited-stage small cell lung cancer. The data from randomized trials suggest that higher doses of thoracic radiotherapy produce better local control and progression-free survival. Although the optimal dose has not yet been established, those trials that demonstrate a superior survival outcome from radiotherapy and chemotherapy over chemotherapy alone have generally used a total dose i

4 of at least 40 Gy in 15 fractions over three weeks (or a biologically equivalent dose). The radiation oncologist must assess the appropriateness and safety of this recommendation for individual patients, taking into consideration tumour field size and location, pulmonary function tests and other clinical factors. These factors are important as the improvement in overall survival occurs with an increased risk of death due to the toxicity of combined modality therapy. There is conflicting evidence as to the optimal timing of thoracic radiotherapy in relation to the course of chemotherapy (early or late administration of thoracic radiotherapy). The evidence is also conflicting regarding the issue of concurrent versus sequential administration of chemotherapy with radiotherapy. Based on currently available data, hyperfractionated thoracic radiotherapy is NOT recommended for limited-stage small cell lung cancer outside of a clinical trial. Methods Entries to MEDLINE and CANCERLIT (through December and October 2002, respectively) and Cochrane Library (through Issue 4, 2002) databases have been searched for evidence relevant to this practice guideline. The most recent literature search was performed in January Evidence was selected and reviewed by one member of the Cancer Care Ontario Practice Guidelines Initiative s Lung Cancer Disease Site Group and methodologists. This practice guideline has been reviewed and approved by the Lung Cancer Disease Site Group, which comprises medical and radiation oncologists, pathologists, surgeons, a respirologist, a medical sociologist, and two community representatives. External review by Ontario practitioners was obtained through a mailed survey. Final approval of the original guideline report was obtained from the Practice Guidelines Coordinating Committee. The Cancer Care Ontario Practice Guidelines Initiative has a formal standardized process to ensure the currency of each guideline report. This consists of periodic review and evaluation of the scientific literature and, where appropriate, integration of this literature with the original guideline information. Key Evidence The Role of Radiotherapy with Chemotherapy Two published meta-analyses compared chemotherapy plus thoracic radiotherapy with chemotherapy alone. The first meta-analysis analyzed published results from 11 trials and the second examined individual patient data from 13 trials; there was substantial overlap between the trials analyzed in the two meta-analyses. Both meta-analyses demonstrated positive benefits for thoracic radiotherapy in combination with chemotherapy versus chemotherapy alone. One meta-analysis demonstrated an overall benefit of thoracic radiotherapy on two-year survival [Odds Ratio, 1.53; 95% Confidence Interval, 1.30 to 1.76; p=0.001] and an absolute improvement in local control of 25.3% (95% Confidence Interval, 16.5 to 34.1). The second meta-analysis indicated a three-year overall survival benefit of 5.4% ± 1.4% (standard deviation) and a Relative Risk of death of 0.86 (95% Confidence Interval, 0.78 to 0.94; p=0.001) in favour of the combined modality group. One of two randomized trials that was not included in either meta-analysis accrued 97 patients and detected a survival benefit for combined modality treatment over chemotherapy alone. The other trial, which involved the use of split-course radiotherapy and a second randomization to consolidation chemotherapy, detected no significant difference in overall survival between treatments among 386 patients, although there was a significant advantage in two-year ii

5 progression-free survival for irradiated patients. The reliability of the results of the latter trial is questionable since the combined treatment arm was closed early due to toxicity. Radiotherapy Timing Concurrent versus Sequential or Alternating Administration Three randomized controlled trials compared concurrent chemo-radiotherapy with either sequential or alternating chemo-radiotherapy. One trial demonstrated a non-significant increase in overall survival for patients receiving thoracic radiotherapy concurrently with chemotherapy versus sequentially following chemotherapy (p=0.097 logrank). However, a regression analysis adjusted for prognostic variables detected a significant survival benefit for concurrent treatment (Hazard Ratio, 0.70; 95% Confidence Interval, 0.52 to 0.94, p=0.02). Another small trial available only in abstract form reported no survival benefit for concurrent over sequential administration of radiotherapy (p=0.33). One randomized controlled trial which compared concurrent chemo-radiotherapy with chemotherapy alternating with thoracic radiotherapy showed no significant difference between the two treatment arms (p=0.15 logrank). Radiotherapy Timing Early versus Late Administration Five randomized controlled trials investigated early versus late thoracic radiotherapy delivery. Methodologists working with the Lung Cancer Disease Site Group conducted a meta-analysis of published data involving 777 patients from three of the randomized controlled trials that examined early versus late daily thoracic radiotherapy delivery. Two of these trials administered chemotherapy concurrently with the radiotherapy and one administered it sequentially. Results of the meta-analysis indicated that there was no survival benefit to administering thoracic radiotherapy early in relation to the chemotherapy administration schedule (Odds Ratio, 1.04; 95% Confidence Interval, 0.45 to 2.43; p=0.9), although the treatment effects detected in the three trials were heterogeneous. Only one of these trials obtained a significant result: the National Cancer Institute of Canada detected a survival advantage for early, concurrent administration of thoracic radiotherapy compared with late, concurrent administration (5-year survival, 20% versus 11%, respectively, p=0.008 log rank). In addition, two randomized controlled trials compared early administration of hyperfractionated thoracic radiotherapy (concurrent with the first course of chemotherapy) to late administration (given concurrently with cycle three or four of chemotherapy). In one of those trials, early administration achieved a significantly higher local control rate and an improvement in survival that was close to statistical significance. In the other trial, there were no differences between administration schedules in complete response rate or survival. Radiotherapy Dosage Two randomized controlled trials examining radiotherapy dosage reported no significant survival benefit of high dose over low dose thoracic radiotherapy; although in one trial, there was an improvement in local control at higher doses. Hyperfractionated Radiotherapy Hyperfractionated thoracic radiotherapy has been shown in one large, fully published study (417 patients) to significantly increase the long-term survival of patients with limited small cell lung cancer (5-year survival, 26% with hyperfractionated thoracic radiotherapy versus 16% with once daily radiotherapy, p=0.04 logrank). This was achieved with an increased rate of short-term grade 3 esophagitis. A second large randomized trial (262 patients) has recently been fully published and has not detected a survival advantage for hyperfractionated thoracic radiotherapy (3-year survival, 29% with hyperfractionated thoracic iii

6 radiotherapy versus 34% with once daily radiotherapy, p=0.49). Grade 3 esophagitis was again significantly more frequent in the hyperfractionated arm. Adverse Effects One meta-analysis demonstrated an increased risk of toxic death in the combined chemotherapy-radiotherapy group compared with the chemotherapy alone group (Odds Ratio, 2.54; 95% Confidence Interval, 1.90 to 3.18; p<0.01). In one trial, grade 3 and 4 thrombocytopenia was increased in the early hyperfractionated radiotherapy group compared with late hyperfractionated radiotherapy (p=0.062). Related Guidelines Cancer Care Ontario Practice Guidelines Initiative s Practice Guideline Reports: : The Role of Combination Chemotherapy in the Initial Management of Limited-Stage Small Cell Lung Cancer : Prophylactic Cranial Irradiation in Small Cell Lung Cancer. Prepared by the Lung Cancer Disease Site Group For further information about this practice guideline report, please contact Dr. William K. Evans, Chair, Lung Cancer Disease Site Group, Cancer Care Ontario, 620 University Avenue, Toronto ON M5G 2L7; TEL (416) ext. 1650; FAX (416) iv

7 PREAMBLE: About Our Practice Guideline Reports The Cancer Care Ontario Practice Guidelines Initiative (CCOPGI) is a project supported by Cancer Care Ontario (CCO) and the Ontario Ministry of Health and Long-Term Care, as part of the Program in Evidence-based Care. The purpose of the Program is to improve outcomes for cancer patients, to assist practitioners to apply the best available research evidence to clinical decisions, and to promote responsible use of health care resources. The core activity of the Program is the development of practice guidelines by multidisciplinary Disease Site Groups of the CCOPGI using the methodology of the Practice Guidelines Development Cycle. 1 The resulting practice guideline reports are convenient and up-to-date sources of the best available evidence on clinical topics, developed through systematic reviews, evidence synthesis and input from a broad community of practitioners. They are intended to promote evidence-based practice. This practice guideline report has been formally approved by the Practice Guidelines Coordinating Committee, whose membership includes oncologists, other health providers, community representatives and Cancer Care Ontario executives. Formal approval of a practice guideline by the Coordinating Committee does not necessarily mean that the practice guideline has been adopted as a practice policy of CCO. The decision to adopt a practice guideline as a practice policy rests with each regional cancer network that is expected to consult with relevant stakeholders, including CCO. Reference: 1 Browman GP, Levine MN, Mohide EA, Hayward RSA, Pritchard KI, Gafni A, et al. The practice guidelines development cycle: a conceptual tool for practice guidelines development and implementation. J Clin Oncol 1995;13(2): For the most current versions of the guideline reports and information about the PEBC, please visit the CCO website at: For more information, contact our office at: Phone: ext Fax: ccopgi@mcmaster.ca Copyright This guideline is copyrighted by Cancer Care Ontario; the guideline and the illustrations herein may not be reproduced without the express written permission of Cancer Care Ontario. Cancer Care Ontario reserves the right at any time, and at its sole discretion, to change or revoke this authorization. Disclaimer Care has been taken in the preparation of the information contained in this document. Nonetheless, any person seeking to apply or consult these guidelines is expected to use independent medical judgment in the context of individual clinical circumstances or seek out the supervision of a qualified clinician. Cancer Care Ontario makes no representation or warranties of any kind whatsoever regarding their content or use or application and disclaims any responsibility for their application or use in any way. v

8 Evidence-based Series #7-13-3: Section 2 The Role of Thoracic Radiotherapy as an Adjunct to Standard Chemotherapy in Limited-Stage Small Cell Lung Cancer: A Systematic Review G. Okawara, A. Gagliardi, W. K. Evans, and members of the Lung Cancer Disease Site Group A Quality Initiative of the Program in Evidence-based Care (PEBC), Cancer Care Ontario (CCO) Report Date: January 2003 I. QUESTION Is there a role for thoracic radiotherapy (TRT) as an adjunct to standard chemotherapy in limited-stage small cell lung cancer (SCLC)? II. CHOICE OF TOPIC AND RATIONALE Combination chemotherapy is the conventional treatment modality for both limited- and extensive-stage small cell lung cancer. However, despite the systemic nature of this malignancy, thoracic control of disease remains problematic. Local failure rates of 80 to 90% have been reported in the literature (1,2). The results of trials conducted to assess the value of thoracic radiotherapy in limited-stage disease have been inconsistent. Possible explanations have been discussed by Arriagada et al (3) and include variations in treatment protocols, nonsignificant statistical variations and lack of sufficient statistical power. Therefore, the Lung Cancer Disease Site Group (Lung DSG) decided to review the evidence as to whether thoracic radiotherapy offers any survival benefit to patients with small cell lung cancer, when thoracic radiotherapy should be delivered with respect to the timing of chemotherapy, and at what dose. III. METHODS Guideline Development This practice guideline report was developed by the Cancer Care Ontario Practice Guidelines Initiative (CCOPGI), using the methodology of the Practice Guidelines Development Cycle (1u). Evidence was selected and reviewed by one member of the CCOPGI s Lung DSG and methodologists. Members of the Lung DSG disclosed potential conflict of interest information. 1

9 The practice guideline report is a convenient and up-to-date source of the best available evidence on the role of thoracic radiotherapy as an adjunct to standard chemotherapy in limitedstage small cell lung cancer, developed through systematic reviews, evidence synthesis, and input from practitioners in Ontario. It is intended to promote evidence-based practice. The Practice Guidelines Initiative is editorially independent of Cancer Care Ontario and the Ontario Ministry of Health and Long-Term Care. External review by Ontario practitioners was obtained through a mailed survey consisting of items that address the quality of the draft practice guideline report and recommendations, and whether the recommendations should serve as a practice guideline. Final approval of the original guideline report was obtained from the Practice Guidelines Coordinating Committee. The CCOPGI has a formal standardized process to ensure the currency of each guideline report. This consists of periodic review and evaluation of the scientific literature and, where appropriate, integration of this literature with the original guideline information. Literature Search Strategy A MEDLINE search was initially conducted from 1990 to July 1998 and updated to March 1999, using the terms [lung neoplasms AND carcinoma, small cell AND thoracic (tw)]. A CANCERLIT search was conducted for 1995 to July 1998 and updated to February 1999, using the same terminology. The Physician Data Query File (PDQ) was also searched for clinical trials using the terms [lung cancer, small cell AND radiation therapy] as was the Cochrane Library (1998, issue 2). The original literature search has been updated using MEDLINE and CANCERLIT (through December and October 2002 respectively), and the Cochrane Library (Issue 4, 2002). The proceedings of the annual meeting of the American Society of Clinical Oncology ( ) were also searched. Inclusion Criteria Articles were selected for inclusion in this systematic review of the evidence if they met the following criteria: 1. Meta-analyses or randomized controlled trials (RCTs) that compared chemotherapy plus radiotherapy with chemotherapy alone, early with late TRT, sequential with concurrent TRT, or different doses of TRT in patients with limited-stage small cell lung cancer. Limited-stage small cell lung cancer is defined as a tumour confined to the hemithorax of origin, the mediastinum and the supraclavicular nodes, which can be encompassed within a "tolerable" radiotherapy port (Physician Data Query (PDQ), National Cancer Institute). Early radiation is generally defined as radiation therapy that is given within the first several cycles of chemotherapy, whereas late radiation therapy is radiotherapy started with the last scheduled course of chemotherapy or after the total course of chemotherapy is completed. Synthesizing the Evidence The Cancer Care Ontario Practice Guidelines Initiative s methodologists pooled five-year survival data from four RCTs comparing early to late TRT to obtain a more precise estimate of the effect of TRT given early or late in the chemotherapy regimen. The Meta-Analyst program provided by Dr. J. Lau, Tufts New England Medical Centre, was used to perform this analysis. Where the data were analyzed by the Cancer Care Ontario Practice Guidelines Initiative, odds ratios and 95% confidence intervals were calculated using a random effects model. Results are expressed such that a mortality odds ratio less than 1.0 favours early TRT. In contrast, the data from one published report of a meta-analysis (4) was reported as the odds ratio of surviving, and in this case, a ratio greater than 1.0 favours early TRT. 2

10 IV. RESULTS Literature Search Results Two published meta-analyses comparing chemotherapy plus TRT with chemotherapy alone (CT) were eligible for review. The first meta-analysis analyzed published results from 11 trials and the second examined individual patient data from 13 trials; there was substantial overlap between the trials analyzed in the two meta-analyses. Nine (RCTs) were also eligible for review (six were fully published). Six of the nine RCTs investigated the timing of TRT delivery. The Cancer Care Ontario Practice Guidelines Initiative s methodologists pooled published data from four RCTs examining early versus late TRT delivery. One of the nine RCTs analyzed optimal dosage of TRT delivered in conjunction with chemotherapy, while two RCTs examined single- versus twice daily TRT treatment in conjunction with chemotherapy. Seven papers identified by a literature search from October 2000 to December 2002 were eligible for inclusion in the systematic review of the evidence (2u-8u). Two of these papers led the Lung Cancer DSG to modify its recommendations in October One paper reported a trial that compared chemotherapy combined with either concurrent or alternating radiotherapy (2u). The other paper was the full report of a trial previously published in abstract form (19) that compared chemotherapy plus daily radiotherapy versus chemotherapy plus twice-daily radiotherapy (3u). Five additional reports of randomized trials were found (4u-8u), one of which included updated results for a trial included in the original guideline report (4u). Outcomes Chemotherapy versus Chemo-Radiotherapy The results of two published meta-analyses which compared chemotherapy alone with combined chemo-radiotherapy are summarized in Table 1. There was substantial overlap between the trials analyzed in each of the meta-analyses; ten of the trials cited in each metaanalysis were common to both. Table 1. Meta-analyses examining effectiveness of thoracic radiotherapy (TRT) plus chemotherapy versus chemotherapy alone. Reference Trials Examined Survival Local Control Warde 11 OR 1.53* (4) 1992 n=1911 (95% CI, 1.30 to 1.76) Pignon (5) 1992 individual patient data 13 n=2140 p< % +/- 1.4% RR 0.86 (95% CI, ) p=0.001 OR 3.02 (95% CI, 2.80 to 3.24) p< Not reported * odds ratio for treatment effect is the odds of surviving two years among patients receiving TRT compared with the odds of patients in the control group absolute increase in overall survival at three years, with a standard deviation of 1.4% local control, odds ratio for treatment benefit, based on nine studies relative risk of death CI confidence interval The Warde analysis of 11 trials reported an improvement in absolute local control of 25.3% (95% confidence interval [CI], 16.5% to 34.1%) for the combined modality group, and a benefit of TRT on two-year survival (odds ratio [OR], 1.53; 95% CI, 1.30 to 1.76; p<0.001) (4). The odds ratio for treatment effect is the odds of surviving two years among patients allocated to receive TRT compared with the odds of surviving for patients allocated to the control group. 3

11 The overall event-rate difference, i.e., the overall benefit on two year survival, was 5.4% (p<0.05; 95% CI, 1.1% to 9.7%). The Pignon analysis of individual patient data from 13 trials showed a similar benefit in favour of the combined modality group (relative risk [RR] of death, 0.86; 95% CI, 0.78 to 0.94; p=0.001) (5) (Note the differences between Warde and Pignon in type and direction of ratios reported; Warde reported an odds ratio of survival, Pignon a relative risk of death). At three years, the absolute increase in overall survival in the group receiving both chemotherapy and thoracic radiotherapy was 5.4% ± 1.4% (standard deviation). Subgroup analysis on the basis of age indicated that the benefit from radiotherapy on mortality was greatest in patients under 55 years of age (p=0.01). The relative risk of death in favour of combination therapy was 0.72 (95% CI, 0.56 to 0.93) for patients under 55 years old and 1.07 (95% CI, 0.70 to 1.64) for patients more than 70 years old; the reductions in the risk of death were 28% + 8% and -7%+17% respectively. The lack of a clearly significant reduction in mortality with radiotherapy found among patients over age 64 years in this subgroup analysis should be interpreted with caution given the wide confidence intervals, the general difficulty in proving a negative result, and the problems inherent in subgroup analysis. Quon et al (6) provide additional data bearing on the issue of treatment effects with age. The authors conducted a retrospective review of data from two previously reported Canadian randomized trials investigating combined modality therapy which included TRT. They partitioned the patients into two age groups, <70 years old and >70 years old. The two groups were compared with respect to baseline patient characteristics, treatment field sizes and doses, numbers of patients receiving and completing TRT, number of days required to complete TRT, toxicity, response rates and survival over the two studies. Analyses indicated that age does not appear to have an impact on the delivery, tolerance or efficacy of TRT for patients with limitedstage small cell lung cancer. Two additional randomized trials of chemotherapy versus chemo-radiotherapy were found during update searches (5u,6u). Johnson et al (5u) compared cyclophosphamidedoxorubicin(adriamycin )-vincristine (CAV) alone with CAV plus concurrent administration of split-course TRT and included a second randomization to consolidation chemotherapy (cisplatin-etoposide) or observation in responding patients. No significant differences were detected between the CAV and CAV-TRT treatment groups with respect to response rate (64% versus 67%, respectively, p=0.58) or overall survival (median, 12.8 versus 14.4 months, respectively, p=0.92 logrank). A significant advantage in two-year progression-free survival was detected for responding patients who received combined modality treatment (39% versus 15%, p=0.002). The authors noted that overall survival was longer with the combined modality treatment (two-year, 33% versus 23.5%, p=0.077; five-year, 16% versus 12%, p=0.36) and suggested that the trial may have been underpowered to detect a modest survival difference. However, the reliability of the trial results are questionable because the combined modality treatment arm was closed early due to toxicity without survival benefit; therefore, randomization was not maintained throughout the study and there was a resulting imbalance in group size (147 received combined modality treatment and 222 received chemotherapy alone). Kraft et al (6u) administered 3 cycles of CAV as an induction regimen and then randomized 97 patients to a) prophylactic cranial irradiation alone (PCI) alone, b) PCI plus TRT at 30 Gy, or c) PCI plus TRT at 50 Gy. Three additional cycles of chemotherapy were administered after completion of TRT. Survival was significantly shorter in the treatment arm without thoracic radiotherapy compared with the two chemotherapy and thoracic radiotherapy treatment arms combined (median, 9.7 versus 12.8 months, p=0.02; two-year, 5% versus 19%, p=0.05); survival did not differ significantly according to radiotherapy dose (median, 13.5 months [30 Gy] versus 12.4 months [50 Gy]; two-year, 25% [30 Gy] versus 13% [50 Gy]). The complete response rate was also higher with the two thoracic radiotherapy treatment arms (both 4

12 47%) compared with PCI alone (26%), although the level of statistical significance was not reported. Toxicity of Chemotherapy versus Combined Modality Therapy The Pignon analysis was unable to evaluate non-lethal treatment toxicity or the incidence of toxic deaths, due to the variable reporting of toxicity by different investigators. Warde calculated the incidence of toxic deaths and found an increased risk of death in the combined modality group. The overall rate difference was 1.2% (95% CI, -0.6% to 3.0%) with a toxic mortality odds ratio (odds of treatment-related deaths in the combined modality group compared with those of the chemotherapy alone group) of 2.54 (95% CI, 1.90 to 3.18; p<0.01). A review by Turrisi (7) acknowledged that the studies analyzed in the two meta-analyses did not employ the chemotherapy regimens most commonly used in 1995, which consisted of platinum-based chemotherapy, and that observed toxicity may have been the result of an interaction between the chemotherapy and radiotherapy. The combined modality regimens employed by both Johnson et al (5u) and Kraft et al (6u) included CAV, although in the former trial administration was concurrent with chemotherapy, while in the latter trial administration was sequential. Grade 4 toxicity was more common with the concurrent, combined treatment (73% versus 56% of patients, p=0.002), which also resulted in more treatment-related deaths (6 versus 3, p=0.1) (5u). Kraft et al reported generally comparable toxicities across treatment arms, although pneumonitis and dysphagia were more common with combined modality treatment even though the treatments were administered sequentially (6u). Timing of Radiotherapy Six RCTs have been published that address the issue of radiotherapy timing. One trial compared the survival of patients receiving radiotherapy either sequentially or concurrently to chemotherapy. Four trials compared the survival of patients receiving either early or late TRT concurrently or sequentially with chemotherapy. Another RCT compared early with late TRT administered twice daily. Details of these trials are summarized in Tables 2a and 2b. Literature review and updating activities identified four trials that addressed the issue of radiotherapy timing (2u,4u,7u,8u). One of these reports (4u) was an update of a trial previously reported in the guideline (8). Details of the trials are included in Tables 2a and 2b. Timing of radiotherapy concurrent versus sequential Although the Pignon meta-analysis (5) examined the question of the timing of TRT (sequential, alternating and concurrent), no statistically significant differences were found among these various treatment schedules. Three out of four trials which demonstrated a significant survival advantage for combined modality therapy employed an alternating or concurrent scheme. Seven out of nine trials which did not demonstrate a survival advantage employed a sequential scheme. Takada et al (8) randomized patients to receive either sequential thoracic radiotherapy or concurrent treatment (published as an abstract). Chemotherapy consisted of cisplatin and etoposide for a total of four cycles. Radiation was initiated after completion of the fourth cycle of chemotherapy in the sequential group, while the concurrent group started thoracic radiotherapy on day 2. The dose employed in both groups was 45 Gy given in a hyperfractionation regimen of 30 fractions over a three-week period. Median survival was 20.8 months for the sequential group and 31.3 months for the concurrent group. Significance levels were not stated. It was acknowledged by the authors that longer follow-up would be required for definitive conclusions. 5

13 A full report of the trial by Takada et al (8) was published in 2002 (4u). Two hundred and twenty-eight eligible patients were recruited between May 1991 and January As of August 2000, median survival times were 19.7 months with sequential TRT and 27.2 months with concurrent therapy (p=0.097 logrank). A regression analysis (Cox model) adjusted for the prognostic factors of age, performance stage, and disease stage suggested a survival benefit for concurrent treatment with a mortality hazard ratio of 0.70 (95 % CI, 0.52 to 0.94, p=0.02). Objective response rates were similar at 92% for sequential treatment and 97% for concurrent treatment, although complete response rates were higher with concurrent therapy (27% versus 40%, p=0.07) In an English abstract of a paper published in Korean, Park et al reported no significant differences between concurrent or sequential chemo-radiotherapy in overall response rates (78% versus 63%, respectively, p=0.13) or survival (mean, 18.3 months versus 13.2 months, respectively, p=0.33) (7u); however, the data available in the abstract report were limited, and with only 51 patients the trial was likely underpowered to detect a survival difference. 6

14 Table 2a. Trials examining timing of thoracic radiotherapy: Treatment regimens. First author, year (reference) Takada 1996 (8) (abstract) 2002 (4u) Lebeau 1999 (2u) Perry 1987 (9) Perry 1998 (10) Perry 1996 (11) Schultz 1988 (12) (abstract) Murray 1993 (13) Work 1997 (14) Jeremic 1997 (15) Treatment Treatment Regimens Groups CT + TRT cisplatin 80mg/m 2 day 1, etoposide 100mg/m 2 days 1-3 q.3-4.wks x3 (sequential) 1.5 Gy bid to 45 Gy over 3 wks after cycle 4 CT + TRT cisplatin, etoposide as above (concurrent) 1.5 Gy bid to 45 Gy over 3 wks beginning day 2, cycle 1 CT + TRT cyclophosphamide 1000mg/m 2 day 1, doxorubicin 45mg/m 2 day 1, (partially etoposide 150mg/m 2 days 1,2, vincristine 1.4mg/m 2 concurrent) with vindesine (3mg/m 2 day 1) replacing doxorubicin for second and third courses of chemotherapy, q4wks 50 Gy in 20 fractions started immediately after second chemotherapy cycle (days 30-64) CT + TRT CT as above (alternating) 20 Gy in 8 fractions starting day 36 (7 days after second chemotherapy cycle) and repeated day 64; 15 Gy in 6 fractions starting day 92 (7 days after fourth chemotherapy cycle). CT cyclophosphamide 1000mg/m 2, etoposide 80mg/m 2, vincristine 1.4mg/m 2 with doxorubicin 50 mg/m 2 replacing etoposide in alternate cycles 7-18 q.3.wks for 18 months CT +TRT CT as above 4000 rad in 4 wks starting day 1, followed by 1000 rad boost after 2 wks CT + delayed CT as above TRT CT + early TRT CT + delayed TRT CT + early TRT CT + late TRT CT + early TRT CT + late TRT CT + early Hfx TRT CT + late Hfx TRT 4000 rad in 4 wks starting day 64 followed by 1000 rad boost after 2 wks cisplatin 60mg/m 2 day 1, etoposide 100mg/m 2 days 4,6,8 alternating with cyclophosphamide 1000mg/m 2, vincristine 1mg/m 2, doxorubicin 45mg/m Gy, 22 fractions CT as above with TRT commencing in the fourth month of treatment cyclophosphamide 1000mg/m 2, doxorubicin 50mg/m 2, vincristine 2mg total dose alternating at 3-week intervals with etoposide 100mg/m 2, cisplatin 25mg/m 2 days 1-3. Each combination administered for three cycles, for a total of six chemotherapy cycles. Third cycle of cyclophosphamide-doxorubicin-vincristine delayed by one week. 40 Gy in 15 fractions over 3 wks concurrent with first cycle of etoposide-cisplatin (week 3) CT as above 40 Gy in 15 fractions over 3 wks concurrent with last cycle of etoposide-cisplatin (week 15) cisplatin 60mg/m 2 + etoposide 120 mg/m 2 for 2 x 3-week cycles followed by cyclophosphamide 1000mg/m 2, doxorubicin 45 mg/m 2 +vincristine 1.4mg/m 2 for 4 cycles, cisplatin-etoposide for 2 cycles, and cyclophosphamide-doxorubicin-vincristine for 2 cycles. TRT of 20 or 22.5 Gy in 11 fractions before and after the first cisplatin-etoposide cycle. CT as above TRT as above but administered before and after the last cisplatin-etoposide cycle (week 18). carboplatin, etoposide 30mg each daily during RT followed by 4 sequential cycles of cisplatin 30mg/m 2 and etoposide 120mg/m 2 days Gy bid to 54 Gy at weeks 1-4 two cycles of cisplatin and etoposide (as above) followed by RT (as above) weeks 6-9 concurrent with carboplatin and etoposide (as above) followed by two more cycles of cisplatin and etoposide cyclophosphamide, doxorubicin, vincristine alternating with etoposide-cisplatin q.3.wks x Gy over 5-6 weeks after cycle 6 CT as above 1.5 Gy bid to 45 Gy concurrent with cycle 1 of etoposide-cisplatin carboplatin AUC 6 day 1 and etoposide 100mg/m 2 days 1-3 q.3.wks x 6 and Park 1996 (7u) CT + TRT (sequential) CT + TRT (concurrent) Skarlos CT + early 2001 (8u) Hfx TRT 1.5 Gy bid to 45 Gy concurrent with cycle 1 CT + late Hfx CT as above and TRT 1.5 Gy bid to 45 Gy concurrent with cycle 4 Notes: bid twice daily, CT chemotherapy, Hfx hyperfractionated, Gy Gray, q every, TRT - thoracic radiotherapy, wk(s) week(s). 7

15 Table 2b. Trials examining timing of thoracic radiotherapy: Results. First author, year (reference) Takada 1996 (8) (abstract) 2002 (4u) Lebeau 1999 (2u) Perry 1987 (9) Perry 1998 (10) Perry 1996 (11) Schultz 1988 (12) (abstract) Murray 1993 (13) Treatment Groups No. Patients* Complete Response Rate CT + TRT (sequential) CT + TRT (concurrent) Sequential Concurrent CT + TRT (partially concurrent) CT + TRT (alternating) CT CT + TRT CT + delayed TRT CT + early TRT CT + delayed TRT % 37.4% As in % 40% p= % 49% 36% 49% 58% 48% 50% Median Survival (months) Overall survival p=0.097 logrank Overall survival: p=0.15 logrank CT + early TRT CT + late TRT % 56% p=0.14 Overall survival: p=0.008 logrank Work 1997 (14) CT + early TRT CT + late TRT % 61% Overall survival: p=0.41 logrank Jeremic 1997 (15) CT + early Hfx TRT CT + late Hfx TRT % 80% p=0.013 Overall survival: p=0.052 Wilcoxon Park 1996 (7u) CT + TRT (sequential) CT + TRT (concurrent) % 30% Mean survival: p=0.33 Wilcoxon Skarlos 2001 (8u) CT + early Hfx TRT CT + late Hfx TRT p=ns p=ns Notes: CT chemotherapy, Hfx hyperfractionated, NR not reported, NS not statistically significant, TRT - thoracic radiotherapy. Timing of radiotherapy concurrent versus alternating Lebeau et al (2u) randomized patients to chemotherapy (cyclophosphamide-doxorubicinvindesine-etoposide-vincristine) combined with either concurrent or alternating radiotherapy. The trial was terminated early, when results from the planned interim analysis indicated a statistically higher mortality rate related to lung toxicity in the concurrent radiotherapy group than in the alternating radiotherapy group (estimated two-year mortality rates, 13% and 2% respectively; p=0.05 logrank). There was no significant difference in overall survival between the groups at the time of the interim analysis. Timing of radiotherapy early versus late In a three-armed study, Perry et al (9) compared patients receiving non-cisplatincontaining chemotherapy alone to chemotherapy combined with either early or late thoracic radiotherapy. Statistically significant advantages with respect to both local control and survival were found in the two groups receiving radiotherapy when compared with the group receiving 8

16 chemotherapy alone. In comparing the two chemo-radiotherapy groups, there was a trend towards improved survival in the late-treatment group when compared with early treatment (p=0.078). In a ten-year update of this study (10) at a time when 95% of the enrolled patients had died, the authors reported that there was a non-significant trend in favour of late versus early radiotherapy (p=0.144). Survival rates at five years were 3% for the chemotherapy-only group, 6.6% for the early-trt group and 12.8% for the late-trt group (11). These results appear substantially worse than the Canadian study reported below by Murray et al (13). The difference may be attributed to: 1) the intentional reduction of chemotherapy doses in the Perry trial for those patients receiving early radiotherapy because of toxicity concerns; and 2) chemotherapy was cyclophosphamide-based in the Perry trial and cisplatin-based in the Murray trial. Schultz (12) randomized patients to receive alternating cycles of chemotherapy consisting of cisplatin/etoposide and cyclophosphamide/doxorubicin/vincristine (published in abstract form), with thoracic radiotherapy started either on day 1 or day 120 and consisted of a dose of at least 40 Gy. Median survival for the early-trt group was 10.7 months compared with 12.9 months for the late-trt group. Results were not statistically significantly different between the two treatment groups. Murray et al (13) reported results of an early versus late thoracic radiotherapy study conducted by the National Cancer Institute of Canada (NCIC). Three hundred and thirty-two patients from 22 centres were entered. Chemotherapy in this study included cisplatin. There were significant improvements in both progression-free survival (p = logrank, Wilcoxon) and overall survival (p = logrank, Wilcoxon) in the early-treatment group. Median survival times for the early and late groups were 21.2 months and 16 months respectively; five-year survival rates were 20% and 11%. The Aarhus Lung Cancer Group published the results of a Danish study (14) which randomized 199 patients to receive early or late radiotherapy. Unlike the Canadian study, this trial did not show improvement in local control (p = 0.2) or overall survival (p = 0.4) for early TRT. Median survival times for the early and late arms were 10.5 and 12.0 months, respectively. These results are worse than the results reported by Murray et al (13) in the Canadian study, as discussed above (21.2 and 16.0 months, respectively). The results of the Danish study may be questioned for a number of reasons. The dose of TRT was changed during the trial; patients treated before October 1984 received 40 Gy and those treated after this date received 45 Gy. In the early-trt group, 40 Gy was the prescribed dose for 45 patients and 45 Gy was the prescribed dose for 54 patients. In the late-trt group, 40 Gy was prescribed to 41 patients and 45 Gy was prescribed to 59 patients. In addition, TRT was prescribed in a split-course manner (two treatment periods of 20 or 22.5 Gy in 11 fractions separated by an interval of 21 days during which chemotherapy was given) which is not radiobiologically optimal. The policy on prophylactic cranial irradiation (PCI) was also modified during the course of treatment. After October 1984, all patients received PCI. Five-year survival data from each of the above four studies comparing early with late TRT were pooled (9-14). There is no survival benefit to administering TRT early in the chemotherapy regimen (odds ratio, 1.00; 95% CI, 0.56 to 1.79; p=1.0). Figure 1 summarizes the data and presents the results of the pooled analysis. Upon reviewing the meta-analysis of early versus late radiotherapy in the original guideline report, two of the four trials appeared to report data from the same trial (12,14). The five-year survival data from the three distinct trials involving 777 patients (9-11,13,14) was pooled using the software package Metaview Software (9u) and a similar result was obtained, indicating no survival benefit for early versus late administration of radiotherapy with chemotherapy (odds ratio, 1.04; 95% CI, 0.45 to 2.43, p=0.9). However, the utility of the pooled effect estimate is questionable since the treatment effects of the individual trials were 9

17 heterogeneous (p=0.028). Reasons for the heterogeneity of trial outcomes are discussed in the original guideline section above. The one trial that detected a significant advantage for early over late radiotherapy, alternated cyclophosphamide-doxorubicin-vincristine with cisplatinetoposide and administered radiotherapy concurrently with cisplatin-etoposide at week 3 (early administration) or week 15 (late administration) (13). The other trial involving concurrent chemoradiation administered radiotherapy on day 1 (early administration) or day 64 (late administration) concurrently with cyclophosphamide-etoposide-vincristine given in three-weekly cycles (9-11). Work et al administered split-course radiotherapy before and after the first (week 3) and last (week 20) cycles of cisplatin-etoposide in a chemotherapy schedule that included a total of three cycles of cisplatin-etoposide and six cycles of cyclophosphamide-doxorubicinvincristine (14). The complete chemotherapy and radiotherapy schedules are shown in Table 2a. Figure 1. Meta-analysis examining early vs. late thoracic radiotherapy. Study Early TRT Late TRT Mortality 95% Confidence Interval Deaths Total Deaths Total Odds Ratio Low High Murray 1993 (13) Perry 1996 (9/10/11) Schultz 1988 (12) Work 1997 (14) TOTAL favours early TRT favours late TRT Timing of radiotherapy hyperfractionated radiotherapy A single RCT using initial versus delayed accelerated hyperfractionated radiotherapy (Hfx TRT) was reported by Jeremic et al (15). A total of 107 patients were randomized between early thoracic radiotherapy (weeks one to four) or late radiotherapy (weeks six to nine). All patients received twice daily treatments of 1.5 Gy to a total dose of 54 Gy. Chemotherapy consisted of cisplatin and etoposide. Median and five-year survival rates were 34 months and 30% for early Hfx TRT, and 26 months and 15% for the delayed group. These results were of borderline significance on univariate analysis (p=0.05) and reached significance with multivariate analysis after controlling for factors such as sex and age (p=0.03). The early-treated group achieved a significantly higher local control rate than the delayed group (p=0.01). The frequency of grade 3 and 4 thrombocytopenia was increased with early Hfx TRT (p=0.06). A new randomized phase II trial by Skarlos et al enrolled 86 patients (8u). All patients in this study received chemotherapy (carboplatin and etoposide) plus concurrent Hfx TRT (45 Gy 10

18 given twice daily in 30 fractions) and were randomized to early (cycle one) or late (cycle four) administration of radiotherapy. There was no significant survival difference between early and late radiotherapy (median, 17.5 months versus 17.0 months, respectively). Doses of Radiotherapy Results of studies of the optimal dose of thoracic radiotherapy are inconclusive. The Danish study employed two dose levels of 40 and 45 Gy (14). The in-field recurrence rate at two years for both groups was 60 to 70%. Two-year overall survival for the 40 Gy group was 15.1% (standard error [SE], 3.8%) compared with 22.1% (SE, 3.9%) for the 45 Gy group. Five-year survival for the 40 Gy group was 9.3% (SE, 3.0%) compared with 12.8% (SE, 3.2%) for the 45 Gy group. Increasing the dose of TRT did not significantly improve overall survival (p=0.18). The NCIC BR3 study randomized limited-stage SCLC patients to two doses of thoracic radiation of 25 and 37.5 Gy (16). Results are summarized in Table 3. There were no reported differences in overall survival or complete response rates between the two groups. Local progression-free survival was improved in the high-dose group (49 weeks, compared with 38 weeks in the low-dose group, p=0.036). Long-term follow-up results have not been reported. Neither study showed any improvement in survival at the higher dose level, but the difference in total dose in the Danish study is small and the Canadian study used doses considered low by current standards. The trial by Kraft et al reported in the Chemotherapy versus Chemo-Radiotherapy section of this report also compared two different doses of radiotherapy, 30 Gy administered over three weeks versus 50 Gy administered over five weeks (6u). They detected no significant differences between the two radiotherapy doses with respect to median survival; two-year survival was 25% with the 30 Gy dose and 13% with the 50 Gy dose. Hyperfractionated Radiotherapy There are two trials that have addressed the issue of hyperfractionated radiotherapy in limited-stage SCLC (Table 3). Turrisi et al (17) published the full report of a large intergroup study, previously reported in abstract form by Johnson et al (18), in which patients were randomized to either daily thoracic irradiation to a total dose of 45 Gy in 25 fractions over five weeks or twice daily treatments of 1.5 Gy in 30 fractions over three weeks. Chemotherapy consisted of 4 cycles of cisplatin/etoposide commencing concurrently with thoracic irradiation. Median survival was 19 months for the daily treated group and 23 months for those treated twice daily. Two year survival rates were 41% and 47% respectively, and 16% and 26% at five years. These results were statistically significant (p=0.04 logrank). The observed rate of local control was higher with twice daily therapy, but this was not significant (local failure rate, 52% with once daily therapy and 36% with twice daily therapy, p=0.06). Toxicity in the form of esophagitis was significantly increased in the group receiving twice daily treatment (63% versus 44%, p<0.001). There were no cases of permanent strictures from acute esophagitis. Bonner et al (19) published an abstract reporting a trial which randomized 253 patients to either once daily treatments to a total dose of 50.4 Gy in 28 fractions or twice daily treatments to a total dose of 48 Gy in 32 fractions given in two courses of 24 Gy each separated by a 2.5 week break. Chemotherapy consisted of six cycles of etoposide and cisplatin with irradiation starting with the fourth cycle in each arm. Median and two-year survival rates were 21 months and 43% for the two arms combined, but the authors did not report response or survival statistics for the individual treatment arms. Grade 3 or greater rates of esophagitis were shown to be significantly increased in the hyperfractionated arm (12.7% versus 4.7%, p=0.027). 11